A Survey on Intelligent Kannada Inscription Character Recognition Using OCR and Machine Learning

Reading a Kannada inscription character in manual is a time-consuming task. Character recognition system using manual takes over a month to identify the character. Over the decades, the character has evolved into different shapes. The archaeology specialists examine each of this character individual...

Full description

Saved in:
Bibliographic Details
Published in:International research journal of innovations in engineering and technology 2024-07, Vol.8 (7), p.154
Main Authors: Shivakumar, B, Asha, K R, Kavyashree, N
Format: Article
Language:English
Subjects:
Archaeology
Identification methods
Inscriptions
Machine learning
Modules
Optical character recognition
Shape recognition
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Reading a Kannada inscription character in manual is a time-consuming task. Character recognition system using manual takes over a month to identify the character. Over the decades, the character has evolved into different shapes. The archaeology specialists examine each of this character individually to identify a character. Manual techniques are inconsistent. Reading the Kannada stone inscriptions character directly would be time-consuming and inefficient task. For both the public and archaeologists, automating the character identification procedure will be advantageous. This is the primary objective of this research, which focused on developing a method to identify ancient Kannada inscription characters using optical character recognition method. The time period from 8 AD to 12 AD was chosen to limit the research scope of the study. The final outcome of the research study has two elements. When a scanned image of the inscription is entered by the user, the OCR module makes it easy way to identify the characters and converted into modern character. The GIS module is used to provide a map for features that track inscription sites and make it easier for users to view the locations of inscriptions. The OCR solution with the best results was further applied after each one was evaluated.
ISSN:2581-3048
DOI:10.47001/IRJIET/2024.807016